GLM 5.3 Flash
High-speed structured extraction & fast tool-calling GLM 5.3 tier.
Fast 355B GLM 5.3 Flash variant built for low-latency JSON extraction, high-frequency tool calls, and rapid multimodal classification.
Preserves Z.ai structured output discipline at substantially lower latency and token draw than full GLM 5.3, ideal for rapid API parsing and agent router steps.
Fast JSON extractionTool routingHigh-frequency classificationRouter classifiersStructured extractionRapid API tool chains
Strengths
Very fast response times, clean JSON tool schemas, vision support, 1M context.
Considerations
Ideal middle layer for schema-validated routing and extraction.
Input$0.06/1M
Output$0.2/1M
Cached read - /1M
Live throughput history
source: status.ebotservers.comParameters355B
Context1M
Max output128K
Intelligence index28
TPS avg 24h0.0
TTFB p500 ms
Uptime100.0%
Call it
Other models from Z.ai
A new version of this page is available